Dry Multisensor Ore Sorting for Pre-Concentration
AI-assisted dry sensor-based ore sorting to upgrade low-grade or diluted feed before downstream processing, including gold stockpiles and spodumene-bearing pegmatite. The workflow uses multisensor classification to reject waste such as schist, improve pre-concentrate quality, reduce reliance on Dense Media Separation, lower water use and operating cost, and improve downstream processing efficiency.
Business Blueprint
GROUNDEDDry multisensor ore sorting upgrades marginal or diluted ore before the plant, rejecting waste early so downstream processing handles higher-value feed.
The Problem
Mining operations with low-grade stockpiles or diluted pegmatite feed lose value when waste rock is carried into downstream processing instead of being rejected early.
Plant and processing managers
Downstream circuits receive diluted feed, making separation less selective and limiting plant efficiency.
Mine and project managers
Ore that exists in the mine or stockpile can remain uneconomic unless it is upgraded before full processing.
Operations cost owners
Conventional DMS-heavy routes can be more expensive than dry sensor-based sorting for suitable ore streams.
Cost of Inaction
Low-grade stockpiles can remain uneconomical to process, while schist dilution continues to constrain concentrate quality and downstream efficiency.
Process Fit
Quality control & inspectionAs-Is
Ore is crushed and screened, then lower-grade or diluted material proceeds toward DMS, CIP, or other downstream processing with too much waste still in the feed. Operators rely on conventional beneficiation routes that may not fully handle close-density waste such as schist or very low-grade stockpiles.
To-Be
A dry multisensor sorting stage is placed after crushing and before downstream concentration. The system classifies individual rocks on the conveyor and ejects waste, so the plant receives a cleaner pre-concentrate and avoids spending downstream capacity, water, and operating cost on material that should have been rejected earlier.
Human Checkpoints
- Confirm ore response through testwork before upgrading from a single-sensor or conventional route. — Metallurgy or project lead
- Set the operating target for what should be rejected as waste versus passed forward as product. — Plant manager or metallurgist
- Review production results against grade, recovery, throughput, and downstream plant performance before expanding the sorting duty. — Operations manager
Systems Touched
Business Cycle
Upstream
- A prepared ore size fraction must be available from crushing and screening before sorting.
- Ore response to multisensor sorting should be tested before committing to the production configuration.
- Mine planning and stockpile management need to identify which low-grade or diluted material is suitable for pre-concentration.
Downstream
- DMS receives cleaner feed with schist removed earlier, improving feed quality and process selectivity.
- Low-grade stockpiles can be converted into valuable ore instead of remaining outside the economic processing envelope.
- Downstream processing capacity is preserved for higher-quality material because waste is ejected before it reaches the plant.
Value Evidence
- Gold gradeINCREASED
At Navachab it has been possible to double the gold grade.
- Operating cost versus DMSREDUCED
dry, sensor-based sorting now runs at 1/4 of the overall operational costs, while doubling throughput.
- ThroughputINCREASED
dry, sensor-based sorting now runs at 1/4 of the overall operational costs, while doubling throughput.
- Low-grade stockpile pre-concentrationIMPROVED
these sorters have preconcentrated over 10 million tons of low-grade stockpiles and turned it into valuable ore
- Sorting plant capacityIMPROVED
the dry sorting plant sorts 200 tons per hour to detect material falling below the CIP plant cut-off grade.
- Schist rejection before DMSIMPROVED
The goal is to remove schist early, typically 5 to 20% of ROM, before the Dense Media Separation (DMS) stage
- Water useREDUCED
- Feed quality and process selectivity before DMSIMPROVED
Adoption Journey
LEVEL 1 — QUICK WIN
Gate: Prove value on a representative low-grade or diluted ore sample, including whether the ore responds to a combination of sensors.
Outcome: A decision-ready business case for whether dry sorting can upgrade the target ore stream before committing plant capital.
LEVEL 3 — ADVANCED
Gate: Prove value across more ore sources, stockpiles, size fractions, or sorter units without losing recovery or downstream stability.
Outcome: Scaled pre-concentration capacity that can treat larger portions of ROM or accumulated stockpiles while keeping downstream plants focused on higher-value feed.
Detailed per-level builds in the solution spectrum below
Risk & Governance
Rejecting payable mineral with the waste stream.
Posture: Require ore-response testwork, set conservative operating targets at commissioning, and review reject samples and recovery impact before widening the sorting envelope.
Applying the sorter to the wrong feed size, ore type, or stockpile and missing the business case.
Posture: Limit production use to approved ore sources and size fractions, then expand only after measured grade, cost, throughput, and downstream results hold up.
Downstream disruption if the sorting stage changes feed volume or composition faster than the plant can absorb.
Posture: Treat the sorter as a controlled pre-concentration stage tied to DMS, CIP, and plant planning, with operating reviews before changing reject thresholds or throughput.
Operating Intelligence
How it works
AI runs the operating engine in real time.
Humans govern policy and overrides.
Measured outcomes feed the optimization loop.
Who is in control at each step
Each column marks the operating owner for that step. AI-led actions sit above the divider, human decisions and feedback loops sit below it.
Step 1
Sense
Step 2
Optimize
Step 3
Coordinate
Step 4
Govern
Step 5
Execute
Step 6
Measure
AI lead
Autonomous execution
Human lead
Approval, override, feedback
AI senses, optimizes, and coordinates in real time. Humans set policy and override when needed. Measurements close the loop.
The Loop
6 steps
Sense
Take in live demand, capacity, and constraint signals.
Optimize
Continuously compute the best next allocation or action.
Coordinate
Push those actions into systems, channels, or teams.
Govern
Humans set policies, objectives, and overrides.
Authority gates · 1
The system may not approve a new sorting recipe, recovery floor, or economic cut point without site metallurgist and operations sign-off. [S1][S2][S3]
Why this step is human
Policy decisions affect the entire operating envelope and require organizational authority to change.
Execute
Run the approved operating loop continuously.
Measure
Measured outcomes feed back into the optimization loop.
1 operating angles mapped
Operational Depth
Technologies
Technologies commonly used in Dry Multisensor Ore Sorting for Pre-Concentration implementations:
Key Players
Companies actively working on Dry Multisensor Ore Sorting for Pre-Concentration solutions:
Real-World Use Cases
AI-assisted dry sensor ore sorting for low-grade gold stockpiles at Navachab
Rocks move on a conveyor, sensors scan them, software separates rocks that are touching, and air jets blow waste away so the remaining material has more gold.
Multisensor ore sorting for spodumene pre-concentration at CBL
Rocks move through a machine that scans each piece with X-rays, cameras, and lasers, then quickly kicks out unwanted schist before the valuable lithium ore goes to the next processing step.